RNA recombination is a major factor driving viral evolution and contributing to new disease outbreaks. Therefore, understanding the mechanism of RNA recombination can help scientists to develop longer lasting antiviral strategies. Tombusviruses are one of the best model RNA viruses to study RNA virus recombination. My goals were to dissect the mechanism of tombusviral RNA recombination. To do so, in my thesis, I describe my results on the roles of (i) the viral replicase and the viral RNA templates; and (ii) the effect of host factors on tombusvirus recombination events. To study the mechanism of RNA recombination without the influence of selection pressure on the emerging recombinants, we developed an in vitro RNA recombination assay based on viral RNA templates and purified viral replicase preparations. Using this in vitro assay, we demonstrated that replicase driven template switching is the mechanism of recombination, whereas RNA ligation seems less likely to be a major mechanism. In addition, we also studied the role of RNA substrates, in more detail. Our results showed that viral replicase preferred to use functional RNA domains in the acceptor RNAs over random switching events. Host factors may also play important roles in RNA recombination. Using yeast as a model system for studying replication and recombination of a tombusvirus replicon, we identified 9 host genes affecting tombusvirus RNA recombination. Separate deletion of five of these genes enhanced generation of novel viral RNA recombinants. Further studies on one of these genes, XRN1, a 5-3 exoribonuclease, indicated that it might be involved in degradation of tombusvirus RNAs. Lack of Xrn1p resulted in accumulation of truncated (partially degraded) replicon RNAs, which became good templates for RNA recombination. To further study Xrn1p, we overexpressed Xrn4p of Arabidopsis thaliana, a functional analogue of the yeast Xrn1p, in Nicotiana benthamiana plants. After superinfecting the Xrn4p-overexpressing N. benthamiana with tombusvirus, truncated tombusvirus genomic and subgenomic RNA1 were observed. Some of the identified tombusvirus variants were infectious in protoplasts and could systemically infected N. benthamiana plants. Overall, this is the first report that a single host gene can affect rapid viral evolution and RNA recombination.

Replication of the nonsegmented, plus-stranded RNA genome of Cucumber necrosis tombusvirus (CNV) requires two essential overlapping viral-coded replication proteins, the p33 replication co-factor and the p92 RNA-dependent RNA polymerase. In my thesis I describe (i) the effect of phosphorylation of p33, (ii) the RNA chaperone-like activity of p33, and (iii) the role of HSP70s a host proteins in the viral replication. To test the effect of phosphorylation on p33 function, I used in vitro phosphorylated p33. I found that phosphorylation inhibited the ability of p33 to bind to the viral RNA. Phosphorylation-mimicking mutations rendered p33 nonfunctional in plant protoplasts and in yeast. Based on these results, I propose that the primary function of phosphorylation of p33 is to regulate its RNA binding capacity, which could affect the assembly of new viral replicase complexes, recruitment of the viral RNA template into replication and/or release of viral RNA from replication. Thus, phosphorylation of p33 might help in switching the role of the viral RNA from replication to other processes, such as viral RNA encapsidation and cell-to-cell movement. Small plus-stranded RNA viruses do not code for RNA helicases that would facilitate the proper folding of viral RNAs during replication. Instead, small RNA viruses might use RNA chaperones for replication as shown here for the p33 replication protein. In vitro experiments demonstrated that the purified recombinant p33 facilitated RNA synthesis on plusstranded and double-stranded (ds)RNA templates up to 5-fold. In addition, p33 rendered dsRNA templates sensitive to single-strand specific S1 nuclease, suggesting that p33 can destabilize highly structured RNA. Altogether, the RNA chaperone activity of p33 might perform similar biological functions to the helicases. SSa a yeast HSP70 found in the viral replication complex and shown to facilitate viral replication (Serva and Nagy, 2006)To dissect the mode of action of SSA in the viral replication I used temperature sensitive and deletion mutants. Both showed miss localization of p33 compared to the wild type. Purified SSA rendered non functional bacterial expressed p92 functional in an in vitro replication assay. SSa might play a role in the transportation and assembly of viral replication proteins.

Carnation Italian ringspot virus (CIRV) is a positive-strand RNA virus that assembles its membrane-bound replication complexes at mitochondria in plant cells. This process is accompanied by extensive inward invagination of the mitochondrial outer membrane, leading to the formation of cytosol-filled spherules, wherein viral RNA synthesis occurs. The mechanism by which CIRV is able to induce spherule formation is unknown, however growing evidence suggests that the host-cell ESCRT (Endosomal Sorting Complex Required for Transport) machinery – a multi-protein complex normally involved in late endosome maturation – may be involved. ESCRT consists of ~30 soluble proteins that form sub-complexes assembled at the late endosomal surface, and function in multivesicular body (MVB) biogenesis. While ESCRT is relatively well characterized in yeasts and mammals, comparably little is known about ESCRT in plants. Hence, as an initial step towards understanding the potential role of ESCRT in CIRV replication, we examined the protein-protein interaction network, subcellular localization, and gene expression profiles of the Arabidopsis thaliana ESCRT components. Overall, the results from these studies suggest that ESCRT organization and function is relatively well conserved in plants compared to other eukaryotes. We also observed that ESCRT is important for CIRV replication, as expression of dominant-negative versions of several key ESCRT components reduced CIRV replication efficiency in plant cells. Moreover, the Arabidopsis ESCRT-I component, Vps23A is recruited from late endosomes to mitochondria in plant cells expressing the CIRV replicase protein, p36, and recruitment of Vps23A was shown to be mediated by sequences located at the N terminus of p36. It was also shown that recruitment of Vp23A to mitochondria by p36 does not require the Ubiquitin E2 Variant domain of Vps23A, which is in contrast to recruitment of ESCRT by retroviruses during viral budding in mammalian cells. Taken together, these results support the hypothesis that CIRV recruits ESCRT by a novel mechanism in order to carry out its replication, a finding that may lend important insight to aspects of normal ESCRT function in plants.